Formation of an Original Database and Development of Innovative Deep Learning Algorithms for Detecting Face Impersonation in Online Exams

The issue related to the risk of identity impersonation, where one person can be replaced by another in online exam surveillance systems, poses challenges. This study focuses on the effectiveness of detecting attempts of identity impersonation through face substitution during online exams, with the aim of ensuring the integrity of assessments. The goal is to develop facial recognition algorithms capable of precisely detecting these impersonations, training them on a tailored database rather than biased generic data. An original database of student faces has been created. An algorithm leveraging advanced deep learning techniques such as depthwise separable convolution has been developed and evaluated on this database. We achieved very high levels of precision, reaching an accuracy rate of 98% in face detection and recognition.

暴露分析模拟系统(EXAMS)及其应用

本文介绍了暴露分析模拟系统(EXAMS)的基本内容,并对其在基础研究、化学品环境风险评价及实际水环境中有机污染物研究等方面的应用作了综述。

On College English Final Exams Within a University

This paper discusses the importance of a well-written college English final exam within a university at the end of each semester, the directive function of such an exam, the valuable informative feedback of the detailed scores, the necessity and importance of informing students of the general format of the exam and types of items as early as possible. It also deals with the importance of formative evaluation, the functions of regular scores, problems in retests that test-makers should pay attention to and the different organizations of test-makers.

Toward Developing a Syllabus-Oriented Computer-Based Question-Banks Software to Support Partially Computerized Exams

Aims: This study aims at designing and implementing syllabus-oriented question-bank system that is capable of producing paper-based exams with multiple forms along with answer keys. The developed software tool is named Χ(Chi)-Pro Milestone and supports four types of questions, namely: Multiple-choice, True/False, Short-Answer and Free-Response Essay questions. The study is motivated by the fact that student number in schools and universities is continuously growing at high, non-linear, and uncontrolled rates. This growth, however, is not accompanied by an equivalent growth of educational resources (mainly: instructors, classrooms, and labs). A direct result of this situation is having relatively large number of students in each classroom. It is observed that providing and using online-examining systems could be intractable and expensive. As an alternative, paper-based exams can be used. One main issue is that manually produced paper-based exams are of low quality because of some human factors such as instability and relatively narrow range of topics [1]. Further, it is observed that instructors usually need to spend a lot of time and energy in composing paper-based exams with multiple forms. Therefore, the use of computers for automatic production of paper-based exams from question banks is becoming more and more important. Methodology: The design and evaluation of X-Pro Milestone are done by considering a basic set of design principles that are based on a list of identified Functional and Non-Functional Requirements. Deriving those requirements is made possible by developing X-Pro Milestone using the Iterative and Incremental model from software engineering domain. Results: We demonstrate that X-Pro Milestone has a number of excellent characteristics compared to the exam-preparation and question banks tools available in market. Some of these characteristics are: ease of use and operation, user-friendly interface and good usability, high security and protection of the question bank-items, high stability, and reliability. Further, X-Pro Milestone makes initiating, maintaining and archiving Question-Banks and produced exams possible. Putting X-Pro Milestone into real use has showed that X-Pro Milestone is easy to be learned and effectively used. We demonstrate that X-Pro Milestone is a cost-effective alternative to online examining systems with more and richer features and with low infrastructure requirements.

ISHD:Intelligent Standing Human Detection of Video Surveillance for the Smart Examination Environment

In the environment of smart examination rooms, it is important to quickly and accurately detect abnormal behavior(human standing) for the construction of a smart campus. Based on deep learning, we propose an intelligentstanding human detection (ISHD) method based on an improved single shot multibox detector to detect thetarget of standing human posture in the scene frame of exam room video surveillance at a specific examinationstage. ISHD combines the MobileNet network in a single shot multibox detector network, improves the posturefeature extractor of a standing person, merges prior knowledge, and introduces transfer learning in the trainingstrategy, which greatly reduces the computation amount, improves the detection accuracy, and reduces the trainingdifficulty. The experiment proves that the model proposed in this paper has a better detection ability for the smalland medium-sized standing human body posture in video test scenes on the EMV-2 dataset.

Compliance of Magnetic Resonance Imaging Examination Requests at the Diagnostic Center of the National Social Security Fund of Conakry

Introduction: MRI is a rapidly growing technique with more and more indica-tions and requests in the Republic of Guinea. Its correct prescription is a guar-antee for the satisfaction of the actors, both prescribers, radiologists and pa-tients. The main objective of this study was to evaluate the compliance of MRI examination requests at the Diagnostic Center of the National Social Se-curity Fund (CNSS) in Conakry. Material and Methods: This was a descriptive cross-sectional study of MRI prescription forms sent to the MRI unit of the CNSS Diagnostic Center from February 1 to May 1, 2021. The 8 compliance criteria established by the French High Authority for Health were used to evaluate the compliance of the examination requests. Results: A total of 7003 examination forms were sent to the facility, including 7% (n = 468) of MRIs. 56.2% of MRI requests were performed by specialists. We observed an overall compliance of 10%. Administrative and clinical compliance were missing in 24% and 38%, respectively. More specifically, the purpose of the examination was not mentioned in 60%, followed by the requesting department in 48.1% and the patient’s age in 35.1%. Conclusion: This study allowed us to highlight the gaps in establishing MRI requests. It would be important to organize an awareness campaign for prescribers on the usefulness of correctly filling an MRI request and to design templates to be filled out by prescribers.

医学生临床实习与考研的关系分析

目的:通过对2021届临床医学专业自主实习和非自主实习学生及上述两组学生中考上研究生和没考上研究生的学生OSCE成绩进行分析比较,分析临床医学专业学生实习与考研的关系,旨在找出各类别学生间的成绩差异,并找出差异存在的原因和解决办法。方法:使用GraphPad Prism 8软件对549名临床医学专业各类别实习学生OSCE考试总成绩的及格率进行描述性分析,对总成绩的平均分进行单因素方差分析,比较各类别学生考试成绩之间的差异。结果:在临床医学专业各类别实习学生中,自主实习学生的考研率(43.80%)显著高于非自主实习学生(26.12%),非自主实习考上研究生的学生成绩平均分[(71.63±5.77)分]最高,自主实习没考上研究生的学生成绩平均分[(67.16±5.53)分]最低,各类别实习学生毕业技能考试成绩存在显著差异。结论:临床医学专业学生的考研和实习不是矛盾对立的,其中主观因素起着决定性作用。

强迫思维症状来访者心理干预个案体会

很多考生在大型考试前头脑中会控制不住的反复出现不相干的想法,影响注意力及心身状态,这是考前焦虑以及脑神经过度疲劳引起的强迫思维。本文通过联合运用森田疗法、认知行为疗法、意象对话技术、正念呼吸放松技术,对实际案例进行心理干预,有效改善来访者强迫思维症状,缓解考前焦虑。通过对咨询过程的呈现和分析,希望能将此方法、思路推广应用。

高考综合改革背景下北京市等级考生物学试卷特点的变化

北京市生物学高中学业水平等级性考试试卷具有“强调系统性、注重整体性、聚焦学科性、坚持创新性、突出导向性”的试题特色,更好地发挥了试卷的功能。

生物学批判性思维测评试题的解析

围绕生物学知识与批判性思维目标,通过引用原题、整合改编、自主编制试题,分析生物学批判性思维测评试题,为批判性思维在生物学教学中的应用提供思路,为生物学批判性思维试题的开发研究提供参考。

主观题自动评判算法研究综述

在教育教学中,试卷评判是教师获取学生知识点掌握情况的重要途径。然而,试题评分是一个耗时的过程,主观题的评判更需要阅卷人认真、投入、细致地审阅,需要耗费大量精力。要减轻教师工作压力,提高主观题评判的效率,基于人工智能的自动评判技术非常重要,其中主观题的自动评判是难点。随着机器学习和深度学习等技术在自然语言处理领域的发展,主观题自动评判技术有了较大进展。文中将主观题分为常规型和开放型两类进行文献梳理,总结主观题自动评价的标准和公开数据集,归纳涉及的方法和技术路线,并对主观题自动评判技术未来的研究方向进行总结和展望。

从中考“提分”到高考“必考”:体育中高考改革基本学理的辨正

当下体育中高考改革取向存在较为严重的学理误区和伦理缺失,其中的学理议题涉及体育中高考改革取向的内在动因和效度,通常作为该项改革的逻辑起点。对相关学理误区予以审视和澄清,可为体育中高考改革的决策、调适,以及学校体育合理定位、体教融合提供理论支持。分析认为:(1)相关改革取向与“体育”难以兼容。基于体育的“考”与“教”具有不同的规定,升学区分性考试内容和要求难以评价、促进学生体育素养和能力的发展,并对学校“体教融合”造成消解。(2)相关改革取向与“健康”难以切合。基于体适能不同类型及发展的参照,“体质健康”主导的考试与“健康”存在偏离;考试相关要求或导向对部分学生造成健康隐患,且对肥胖、近视等“健康问题”缺乏针对性。(3)体育中高考改革决策需厘清体育中高考评价的实质、“体育与健康”的定位及改革的利弊,也即明确体育中高考所评价的主要是一种非基本或特定的运动能力。研究提出:体育中高考相关改革取向具有某些结构性的局限,不宜强行推行;作为权宜性举措,应将体育中考明确为“水平考”或“达标考”;对体育中高考相关问题的审视及改革决策,需置于学校体育内在规定、体育“人文-教育”价值与健康追求3要素构成的分析框架之下。

基于全过程评价的混合式一流课程妇产科学教学实践及效果分析

科学的课程考核方案既是评价学生学业水平的重要指标,也是课程持续建设改进的客观依据。本研究以妇产科学为例,通过构建全过程考核评价体系,对学生的学业表现进行全面分析,探究混合式课程的教学效果,以期为新时期高校混合式课程建设及课程考核体系构建提供参考。

团体音乐治疗对大学生考试焦虑的干预研究

考试焦虑是当前高校学生中常见心理问题之一,对大学生的学业成绩与身心健康皆产生重要影响。旨在探讨团体音乐治疗对大学生考试焦虑的干预效果。实验选取存在考试焦虑的大学生30名,随机分入实验组与控制组,通过综合运用接受式音乐治疗、再创造式音乐治疗与即兴演奏式音乐治疗3种方式,对实验组进行为期4周的团体音乐治疗。结果表明,团体音乐治疗对大学生考试焦虑具有良好的干预效果;且团体音乐治疗对大学生考试焦虑的干预效果具有一定的持续性。

体育高考生术科考试焦虑的潜在类别及影响因素研究

目的:探索体育高考生术科考试焦虑的类型及影响因素。方法:以2022年江西省785位体育高考生为调查对象,采用体育术科考试焦虑量表、成就动机量表和自我效能量表进行测量,用潜在类别模型对体育高考术科考试焦虑进行潜在类别探索,用无序多元logistic回归模型分析其影响因素。结果:①《体育高考术科考试焦虑量表》有良好的信效度,可用于体育高考生的专业考试焦虑测量。②体育高考生术科考试焦虑可分三个潜在类别,分别是认知状态焦虑组(11.3%)、躯体和状态自信焦虑组(51.2%)和高焦虑水平组(37.5%);③女生、城市考生和高自我效能者更可能进入认知状态焦虑组,成就动机对高焦虑水平组有显著的预测效应,三组之间复读生的比例没有显著性差异。结论:体育高考生术科考试焦虑有显著的群体异质和个体差异,自我效能、性别和生源地对考生的类别归属有显著影响,高成就动机可缓解高焦虑水平,是否复读对考生的类别归属没有显著影响。

非标准化考试教学改革在中医眼科学的探索与实践

目的探索中医眼科学非标准化考试模式的方法与效果。方法对中医学本科专业学生实行非标准化考试的教学改革,贯穿翻转课堂,启发式讲授、病案式导入等教学方式,采用多元化的形成性评价形式,结合终结性评价模式。结果参与非标准化教学改革的学生90.1%对充分肯定教师教学形式及考核方法的合理性及有效性,95.5%对中医眼科学的教学总体满意度积极肯定。结论中医眼科学的非标准化考试教学改革有利于更加合理、全面的评价教师与学生的教与学成果,提升高等中医学教学质量。

江西省2018—2020年口腔医学专业住院医师规范化培训结业考核成绩分析

目的通过对江西省口腔医学专业住院医师规范化培训(简称住培)结业考核临床实践技能考核成绩进行分析,发现口腔医学专业住培中出现的问题,并给出针对性建议,提升江西省口腔医学专业住培质量。方法选取南昌大学附属口腔医院2018至2020年期间参与住培结业临床实践技能考核的223名口腔医学专业住院医师的考核成绩,根据住院医师的不同身份类型、培训基地、培训专业、性别等特征进行分组,运用SPSS 24.0统计软件录入数据并进行描述性分析,计数资料单因素分析使用χ^(2)检验,计量资料单因素分析使用t检验和方差分析,以分析比较不同住院医师间结业考核成绩的差异。结果2018—2020年江西省口腔结业考核总体平均通过率为92.73%,不同人员类型住院医师考核通过率有显著性差异(P<0.05);不同身份类型、不同培训基地、不同培训专业间住院医师的住培实践技能结业考核平均成绩无显著性差异(P>0.05),不同性别住院医师考核平均成绩无显著性差异(P>0.05),口腔医学硕士专业学位研究生的平均成绩高于其他类型住院医师(P<0.05),省级医院住院医师平均成绩高于市级医院住院医师(P<0.05);不同专业住院医师在第一站和第三站成绩不同(P<0.05)。结论江西省口腔医学专业住培考核整体情况较好,但仍然存在不足,口腔医学专业需更重视临床实践技能的培养,应当优化培养方案和考核标准,严格把控培训的科学性和专业培养的重点,进一步完善现行培养模式及考核制度,培养出更多优秀口腔医学人才。

线上线下体验式教学在护理学导论课程思政中的应用

目的探讨线上线下体验式教学在护理学导论课程思政中的应用效果。方法采用便利抽样法,选取2022级护理本科两个班为试验组,应用线上线下体验式课程思政教学;2021级护理本科两个班为对照组,应用传统课程思政教学。课程结束后比较两组学生的理论考核成绩、课程思政效果及学生对授课模式及提高综合素质的满意度。结果授课结束后,试验组学生理论考核成绩为(83.83±5.84)分,显著高于对照组学生成绩(80.94±5.56)分(P<0.05);试验组对课程思政效果评价明显的比例,对授课模式及提高综合素质的满意度显著高于对照组(均P<0.05)。结论线上线下体验式教学能明显提高课程思政效果,有利于提高学生的综合素质。

虚拟仿真及情景模拟教学法在医学生心电图教学中的应用

目的:探讨虚拟仿真及情景模拟教学法在医学生心电图教学中的应用价值。方法:选取2020~2021年在北京朝阳医院接受培训的30名本科五年制医学生,随机分为两组。试验组(n=15)采用虚拟仿真与情景模拟教学法,对照组(n=15)采用传统教学法。对两组学生进行临床病例分析考试。结果:试验组考试成绩显著高于对照组(P<0.01)。结论:虚拟仿真及情景模拟教学法能够有效激发医学生学习心电图的兴趣和积极性,提高其专业知识掌握和实际问题解决能力。

北京市妇产科住院医师规范化培训不同阶段结业临床技能考核结果分析

目的通过分析比较2016年—2021年北京市妇产科住院医师规范化培训(简称住培)一阶段和二阶段结业技能考核成绩,为提升妇产科住培提供依据。方法比较2016年—2021年北京市住院医师一阶段1011人和二阶段675人结业临床技能考核通过率及单项分值差异。结果一阶段每年及总的通过率均高于二阶段。辅助检查成绩对于一阶段考核的通过率影响最大,而临床技能成绩对于二阶段考核的通过率影响最大。二阶段妇科和产科单项考核成绩存在严重偏科。结论一阶段应重点强化辅助检查方面的培训,二阶段应加强各亚专科临床能力的综合培训。